Getting things in order, AI and perceptrons: Bitcoin Masterclasses #4 with Dr. Craig Wright
We’ve talked about using nLocktime in Bitcoin payment channels, but how do we start ordering all these things? “We have to think a little differently than the typical cypherpunk stuff,” replied Dr. Craig Wright.
Dr. Wright gave a “path based session” tutorial to kick off Day 2 of the latest The Bitcoin Masterclasses series in London last week.
“We don’t want to start thinking that all intermediaries are bad,” says Dr. Wright. His Bitcoin white paper discusses eliminating the need for trusted third parties in small day-to-day payments, but other intermediaries still have points and purposes. That’s especially true when you’re constructing complex payments with locked-in times, fees, and automation to consider.
We could have such an intermediary that “acts like a miner” but manages locked and timed transactions, ensuring that ordered transactions arrive on the network at the right time. Dr Wright draws a diagram of a “very simple perceptron”, where different addresses are effectively inputs into a state machine.
“An address is not what people think. It’s a template.” Bitcoin public keys are just one use for them, and Dr. Wright expected others to explore more. (Trivia: The “1” at the start of many Bitcoin public keys is a reference to “template type one.”)
Do whatever you want before the transaction is complete
We can have multiple potential inputs and multiple potential outputs, and any of these can change before nLocktime expires and a completed transaction goes to the blockchain. A number of linked payment channels can communicate with each other in ways that determine the final (recorded) outcome. We can outsource many of these functions to the market, i.e. third-party providers who develop optimized solutions.
“Apart from chatbots, this is a real use for AI. This is something that has real value,” Dr. Wright pointed out.
So we have these third-party intermediaries that figure out which transactions to send to miners. How do you ensure that they (and the miners) are all being honest with each other and processing the correct payments? Dr Wright also takes a brief foray into the world of incentives, saying that the rules of competition tend to favor honesty, as everyone in an industry is incentivized to identify faults among their competitors – in the hope of knocking them out of the competition.
An example he gives is a hypothetical world where all tickets – concert or plane seats – could be tokenized. Tickets not sold immediately before an event or flight may be offered in a fire sale auction. Discounted tickets may come with special conditions or restrictions. Purchased tickets can even be “scalped” (ahem, legally resold) in a way that still pays a percentage royalty to the original seller. As long as transactions are finally settled, the transaction parties are happy and can still make money, all is well.
When all the steps in a process are in a certain order/sequence, and there is a way to ensure that a transaction is settled, when all the variables in a sequence are sorted out; we can use them for any trading area. Even a global settlement system like SWIFT could be better if banks could account for each payment and the exact order in which they were made.
“You can say ‘this is complex’ – and yes, it is. That’s why we have computers,” Dr Wright said.
Like Bitcoin, the Masterclasses series is Dr. Wright throwing ideas out there and letting smart people come up with innovative ways to monetize them. These classes are worth watching for inspiration alone, and there are plenty of suggestions for the next big thing in each session.
See: Bitcoin Masterclasses Workshop on ideas for implementing Bitcoin blockchain in businesses
New to Bitcoin? Check out CoinGeeks Bitcoin for beginners section, the ultimate resource guide for learning more about Bitcoin – as originally envisioned by Satoshi Nakamoto – and blockchain.